Covariate adjustment in family-based association studies

Department of Epidemiology, Columbia University, New York, New York, United States
Genetic Epidemiology (Impact Factor: 2.6). 04/2005; 28(3):244-55. DOI: 10.1002/gepi.20055
Source: PubMed


Family-based tests of association between a candidate locus and a disease evaluate how often a variant allele at the locus is transmitted from parents to offspring. These tests assume that in the absence of association, an affected offspring is equally likely to have inherited either one of the two homologous alleles carried by a parent. However, transmission distortion was documented in families in which the offspring are unselected for phenotype. Moreover, if offspring genotypes are associated with a risk factor for the disease, transmission distortion to affected offspring can occur in the absence of a causal relation between gene and disease risk. We discuss the appropriateness of adjusting for established risk factors when evaluating association in family-based studies. We present methods for adjusting the transmission/disequilibrium test for risk factors when warranted, and we apply them to data on CYP19 (aromatase) genotypes in nuclear families with multiple cases of breast cancer. Simulations show that when genotypes are correlated with risk factors, the unadjusted test statistics have inflated size, while the adjusted ones do not. The covariate-adjusted tests are less powerful than the unadjusted ones, suggesting the need to check the relationship between genotypes and known risk factors to verify that adjustment is needed. The adjusted tests are most useful for data containing a large proportion of families that lack disease-discordant sibships, i.e., data for which multiple logistic regression of matched sibships would have little power. Software for performing the covariate-adjusted tests is available at

4 Reads
  • Source
    • "Identification of genetic factors which play a role in complex traits is hampered by the fact that phenotypes may be a result of different underlying biological mechanisms. Adjusting the phenotypes for covariates provides often only a small improvement of the efficiency of a test statistic (Whittemore et al., 2005). Modelling interaction between genetic factors and covariates may be more promising. "
    [Show abstract] [Hide abstract]
    ABSTRACT: In order to study family-based association in the presence of linkage, we extend a generalized linear mixed model proposed for genetic linkage analysis (Lebrec and van Houwelingen (2007), Human Heredity 64, 5-15) by adding a genotypic effect to the mean. The corresponding score test is a weighted family-based association tests statistic, where the weight depends on the linkage effect and on other genetic and shared environmental effects. For testing of genetic association in the presence of gene-covariate interaction, we propose a linear regression method where the family-specific score statistic is regressed on family-specific covariates. Both statistics are straightforward to compute. Simulation results show that adjusting the weight for the within-family variance structure may be a powerful approach in the presence of environmental effects. The test statistic for genetic association in the presence of gene-covariate interaction improved the power for detecting association. For illustration, we analyze the rheumatoid arthritis data from GAW15. Adjusting for smoking and anti-cyclic citrullinated peptide increased the significance of the association with the DR locus.
    Biometrical Journal 02/2010; 52(1):22-33. DOI:10.1002/bimj.200900057 · 0.95 Impact Factor
  • Source
    • "It is further noted that model (1) not only encompasses all the DT's, but most importantly it handles covariates directly. The resultant test statistic may be seen as an alternative to the covariate adjusted TDT (Whittemore et al. 2005). As an illustration for a matched case-control design , consider the hypothetical example presented by Spielman & Ewens (1998). "
    GY Zou ·
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper applies a retrospective logistic regression model (Prentice, 1976) using a sandwich variance estimator (White, 1982; Zeger et al. 1985) to genetic association studies in which alleles are treated as dependent variables. The validity of switching the positions of allele and trait variables in the regression model is ensured by the invariance property of the odds ratio. The approach is shown to be able to accommodate many commonly seen designs, matched or unmatched alike, having either binary or quantitative traits. The resultant score statistic has potentially higher power than those that have previously appeared in the genetics literature. As a regression model in general, this approach may also be applied to incorporate covariates. Numerical examples implemented with standard software are presented.
    Annals of Human Genetics 04/2006; 70(Pt 2):262-76. DOI:10.1111/j.1529-8817.2005.00213.x · 2.21 Impact Factor
  • Source
    • "The evidence of an increased risk in relation to the CYP19 (TTTA)11 allele was also observed in the conditional logistic regression analysis adjusting for potential confounding variables among the subset of families containing discordant sibships. However, because of the reduced power of these analyses among only a subset of families [36], results of these discordant sibship analyses did not achieve statistical significance. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Case-control studies have reported inconsistent results concerning breast cancer risk and polymorphisms in genes that control endogenous estrogen biosynthesis. We report findings from the first family-based association study examining associations between female breast cancer risk and polymorphisms in two key estrogen-biosynthesis genes CYP17 (T-->C promoter polymorphism) and CYP19 (TTTA repeat polymorphism). We conducted the study among 278 nuclear families containing one or more daughters with breast cancer, with a total of 1123 family members (702 with available constitutional DNA and questionnaire data and 421 without them). These nuclear families were selected from breast cancer families participating in the Metropolitan New York Registry, one of the six centers of the National Cancer Institute's Breast Cancer Family Registry. We used likelihood-based statistical methods to examine allelic associations. We found the CYP19 allele with 11 TTTA repeats to be associated with breast cancer risk in these families. We also found that maternal (but not paternal) carrier status of CYP19 alleles with 11 repeats tended to be associated with breast cancer risk in daughters (independently of the daughters' own genotype), suggesting a possible in utero effect of CYP19. We found no association of a woman's breast cancer risk either with her own or with her mother's CYP17 genotype. This family-based study indicates that a woman's personal and maternal carrier status of CYP19 11 TTTA repeat allele might be related to increased breast cancer risk. However, because this is the first study to report an association between CYP19 11 TTTA repeat allele and breast cancer, and because multiple comparisons have been made, the associations should be interpreted with caution and need confirmation in future family-based studies.
    Breast cancer research: BCR 01/2005; 7(1):R71-81. DOI:10.1186/bcr951 · 5.49 Impact Factor
Show more

Similar Publications